目录
1 绪论.................................................................1
1.1 研究背景及意义..................................................1
1.2 研究内容........................................................2
1.3 本章小结........................................................2
2 环境搭建.............................................................3
2.1 基础开发环境搭建................................................3
2.1.1 CUDA 的下载安装 ..........................................3
2.1.2 cudnn 的下载安装 ...........................................4
2.2 深度学习环境搭建................................................4
2.2.1 miniconda 的安装............................................4
2.2.2 环境初创建.................................................5
2.3 本章小结........................................................9
3 数据清洗............................................................10
3.1 数据分析.......................................................10
3.2 定义清洗规则...................................................13
3.3 数据清洗.......................................................13
3.4 本章小结.......................................................17
4 深度神经网络模型....................................................18
4.1 CNN 网络结构概述 ..............................................18
4.1.1 人工智能网络..............................................18
4.1.2 卷积神经的层级结构........................................20
4.1.3 卷积计算层................................................21
4.1.4 激励层与池化层............................................22
4.2 LSTM 网络结构概述 .............................................23
4.2.1 RNN 循环神经网络 .........................................24
4.2.2 LSTM 神经网络 ............................................24
4.2.3 LSTM 的核心思想 ..........................................26
4.2.4 LSTM 的逐步解析 ..........................................26
